Always negotiate, sometimes cooperate: an agent-based model

Purpose The purpose of this paper is to use agent-based modeling to simulate the negotiation and cooperation between agents working on tasks in an organization and to study the effects that gender differences might have on the outcome of the process. Design/methodology/approach The model used herein allows for idiosyncratic differences in terms of the propensity to negotiate/cooperate. The model also allows for multi-round negotiations/cooperation and incorporates subjective value into the negotiation process. The model is implemented in NetLogo. Findings The results clearly show that it is always beneficial to negotiate, even when backlash might result from the request. The study then extends this analysis by allowing for gender differences in both negotiation and cooperation. The results provide strong support for the hypothesis that agentic characteristics are beneficial for negotiators, while communal characteristics can be detrimental. Research limitations/implications Like all models, the model used herein made some simplifying assumptions about the negotiation and cooperation processes. In addition, the utilized model assumes that agents work individually on tasks and that negotiation takes place between two individual agents, even though negotiation can be a team-based endeavor in many cases. Practical implications The results of this study indicate that individuals need to adopt characteristics that are more agentic; this finding is particularly true for females who aim to be on a level playing field with their male counterparts. The results also indicate that negotiation is beneficial whether there is an abundance of resources or not, while cooperation is only beneficial when resources are abundant. Originality/value While past negotiation research has used simple choice games, laboratory studies and field studies, this study provides computational support for the hypothesis that higher levels of negotiation are beneficial to individuals. Additionally, unlike recent agent-based studies that have studied negotiation as a taken-for-granted automated computational process that is done by software on behalf of individuals, the present study simulates agents that have yet to decide whether they will engage in negotiation or not.